QUESTIONS UNDER STUDY: The optimal strategy of anti-infectious prophylaxis in patients with acute leukaemia undergoing intensive chemotherapy remains a matter of debate. We assessed the impact of primary prophylaxis with posaconazole and levofloxacin on the incidence of invasive fungal infections (IFI) and bacteraemia. METHODS: A retrospective single-centre study including two groups of adult patients with AML receiving intensive chemotherapy. Group one without anti-infective prophylaxis (September 2008 - February 2010), and group two with anti-infective prophyalaxis (March 2010 - April 2011). The primary end-point was IFI according to the EORTC/MSG 2008 definitions and bacteraemia. RESULTS: Baseline characteristics were similar in the non-prophylaxis (n = 43 patients; 99 chemotherapy cycles) and the prophylaxis (n = 45; 104 chemotherapy cycles) group. IFI were significantly reduced in the prophylaxis group (55.3% vs. 88.9%; p = 0.0032) and there was a trend of the projected IFI-free survival at 100 days to be increased (50.1% vs. 25%; p = 0.0526). One-hundred day overall survival (84.4% and 88.4%, p = 0.35) and 2-year overall survival (64.4% and 58.1%; p = 0.64) were unaffected. No difference in the occurrence of bacteraemia was observed (32.3% vs. 34.6%; p = 0.8). A total of two (3.6%) patients in the non-prophylaxis and three (6.7%) in the prophylaxis group died due to IFI, and two (3.6%) in the non-prophylaxis and none in the prophylaxis group patients had to stop leukaemia treatment due to IFI. CONCLUSIONS: The anti-infective prophylaxis with posaconazole and levofloxacin resulted in a significant reduction of 'possible' IFI with a number-needed to treat to prevent one IFI of only 3 but did not result in a reduction of the incidence of bacteraemia.